Multi-provider model workflows will deliver the next math breakthroughs, predicts Andrew Critch
sebkrier · x · 2026-11-08
Andrew Critch predicts independent teams combining frontier models from two or more providers will start producing stronger math results — citing Elliot Glazer's proof that ℤ is Diophantine over ℚ, achieved with Gemini (Astra) and Claude, via a polynomial with 292.9 billion variables. He expects basic physics and materials science to follow: single-provider agent swarms get the hype first, but cross-provider combinations will win on results per dollar — and won't be advertised, since no single lab funds that marketing.
More from AGI Musings
- Shopify CEO Tobi Lütke at DotDev 2026: culture is how much a company sweats the details — schwentker · 2026-11-08
- Tobi Lütke at DotDev 2026: entrepreneurship is the most AI-safe job in the world — schwentker · 2026-11-08
- Locally deployed LLM extracts 136K data points from 130K clinical notes with 97.5% accuracy — DrDatta_AIIMS · 2026-11-08
- Why women trust AI less: missing voices, extra cleanup work, competence penalties — AnnaCiaunica · 2026-11-08
- It's 2030: burnout replaced by boredom, performance reviews by 'what did I accomplish?' — YogeshMalik · 2026-11-08
- Addy Osmani: Managing coding agents is just management — and it beats prompt engineering — addyosmani · 2026-11-08